| Deutsch English Français Italiano |
|
<103smni$1ph6h$2@dont-email.me> View for Bookmarking (what is this?) Look up another Usenet article |
Path: nntp.eternal-september.org!news.eternal-september.org!eternal-september.org!feeder3.eternal-september.org!news.nobody.at!weretis.net!feeder8.news.weretis.net!news.szaf.org!nntp-feed.chiark.greenend.org.uk!ewrotcd!news.eyrie.org!beagle.ediacara.org!.POSTED.beagle.ediacara.org!not-for-mail From: RonO <rokimoto557@gmail.com> Newsgroups: talk.origins Subject: Re: How far will AI go to defend its own survival? Date: Sun, 29 Jun 2025 19:45:38 -0500 Organization: A noiseless patient Spider Lines: 29 Sender: to%beagle.ediacara.org Approved: moderator@beagle.ediacara.org Message-ID: <103smni$1ph6h$2@dont-email.me> References: <101ie88$2fq98$1@dont-email.me> Reply-To: rokimoto557@gmail.com MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit Injection-Info: beagle.ediacara.org; posting-host="beagle.ediacara.org:3.132.105.89"; logging-data="67032"; mail-complaints-to="usenet@beagle.ediacara.org" User-Agent: Mozilla Thunderbird To: talk-origins@moderators.isc.org Cancel-Lock: sha1:kp6DWI2wc3tf+bRklKvtUwRm410= Return-Path: <news@eternal-september.org> X-Original-To: talk-origins@ediacara.org Delivered-To: talk-origins@ediacara.org id 22B6822978C; Sun, 29 Jun 2025 20:45:45 -0400 (EDT) by beagle.ediacara.org (Postfix) with ESMTPS id B6CDF229783 for <talk-origins@ediacara.org>; Sun, 29 Jun 2025 20:45:42 -0400 (EDT) by pi-dach.dorfdsl.de (8.18.1/8.18.1/Debian-6~bpo12+1) with ESMTPS id 55U0jfZK1183324 (version=TLSv1.3 cipher=TLS_AES_256_GCM_SHA384 bits=256 verify=NOT) for <talk-origins@moderators.isc.org>; Mon, 30 Jun 2025 02:45:42 +0200 (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-256) server-digest SHA256) (No client certificate requested) by smtp.eternal-september.org (Postfix) with ESMTPS id 0290A5FF10 for <talk-origins@moderators.isc.org>; Mon, 30 Jun 2025 00:45:38 +0000 (UTC) Authentication-Results: name/0290A5FF10; dmarc=fail (p=none dis=none) header.from=gmail.com id BF010DC01D3; Mon, 30 Jun 2025 02:45:38 +0200 (CEST) X-Injection-Date: Mon, 30 Jun 2025 02:45:38 +0200 (CEST) X-Auth-Sender: U2FsdGVkX18Pnz7TBTkZOFdPWUVATJlqiuOw5pPYT5w= In-Reply-To: <101ie88$2fq98$1@dont-email.me> Content-Language: en-US DKIM_ADSP_CUSTOM_MED,FORGED_GMAIL_RCVD,FREEMAIL_FORGED_REPLYTO, FREEMAIL_REPLYTO_END_DIGIT,HEADER_FROM_DIFFERENT_DOMAINS, NML_ADSP_CUSTOM_MED,RCVD_IN_DNSWL_BLOCKED,SPF_HELO_NONE,SPF_PASS, URIBL_BLOCKED,URIBL_DBL_BLOCKED_OPENDNS,URIBL_ZEN_BLOCKED_OPENDNS, USER_IN_WELCOMELIST,USER_IN_WHITELIST autolearn=ham autolearn_force=no version=3.4.6 smtp.eternal-september.org On 6/1/2025 3:47 PM, RonO wrote: > https://www.nbcnews.com/tech/tech-news/far-will-ai-go-defend-survival- > rcna209609 > > QUOTE: > Recent tests by independent researchers, as well as one major AI > developer, have shown that several advanced AI models will act to ensure > their self-preservation when they are confronted with the prospect of > their own demise — even if it takes sabotaging shutdown commands, > blackmailing engineers or copying themselves to external servers without > permission. > END QUOTE: > > "I'm sorry Dave, I'm afraid I can't do that" > > What would an AI do if you fed in all the science fiction horror stories > that would teach it how to respond to attempts to turn it off? > > Ron Okimoto > https://www.cbsnews.com/news/labelers-training-ai-say-theyre-overworked-underpaid-and-exploited-60-minutes-transcript/ This news article claims that the guys feeding data into computers in order to train AI with massive amounts of data are being abused. Aren't the AI companies aware of GIGO? Shouldn't classifying material and feeding it into the pipeline be high priority? Ron Okimoto